The Pacific Landfalling Jets Experiment (PACJET)
and a Long-term Effort to Improve 0-24 h West Coast Forecasts

27 January 2000

(Contact: Dr. Martin Ralph, NOAA/Environmental Technology Lab., Marty.Ralph@noaa.gov)

1. Introduction

a) Overarching goal

To develop and test methods to improve short-term (0-24 h) forecasts of damaging weather on the U. S. West Coast in landfalling winter storms emerging from the data sparse Pacific Ocean. PACJET is designed to be a next step toward attaining this goal. It is envisioned as part of a long-term effort that combines analysis, a series of focused field experiments, development of new products and tools for operational forecasting, and exploration of physical processes that contribute to the linkage between seasonal-to-interannual climate variability (e.g., ENSO) and extreme coastal weather events.

b) Time and location of PACJET

10 January 2001 to 5 March 2001

From 300 km inland to 1000 km offshore of the U.S. West Coast from Southern California to Washington State. This covers the area between the coast and the central Pacific domain sampled by the Winter Storm Reconnaissance program (Figs. 1 and 2).

Fig. 1. Base map showing anticipated flight area for the PACJET experiment (bold outline) to be conducted from 10 January to 5 March 2001. Flights will be operated mostly out of Monterey, CA, but some will be staged out of Seattle, WA. The domain of an earlier pilot project, CALJET, is shown for reference, as are the flight areas associated with upper-level synoptic-scale, aircraft targeting missions flown out of Hawaii and Alaska as part of NCEP s Winter Storms Program (WSP).

Fig. 2. Base map showing PACJET sampling areas. Aircraft and ship operations areas offshore are shown, as are coastal areas of special interest for physical process studies. One of these coastal areas will be selected based on seasonal-to-interannual (ENSO) forecasts. A region in California that will be heavily sampled during a concurrent air quality study, including 20 upper-air sites, is also highlighted. Coastal wind profiler sites are shown, as is the location of a prototype buoy-mounted wind profiler.

c) Background

PACJET addresses short-term weather prediction along the US West Coast during winter, focusing on quantitative precipitation forecasting. It builds on experience gained from CALJET, an earlier west coast experiment performed during the strong El Niño of 1997/98. CALJET aided flood forecasting (Fig. 3), studied the utility of new observations, and explored the relationship between El Niño and extreme coastal storms. PACJET expands the area covered by CALJET and targets the coastal zone, which is an area of extreme societal vulnerability for which forecasts are typically less accurate than normal, and thus is a problem of great interest to NOAA. PACJET directly links core USWRP objectives on quantitative precipitation forecasting (QPF) and Optimal Observing Systems to this difficult but important problem.

Click for full sized figure.

Fig. 3. Summary of the storm of 2-3 February 1998 observed during CALJET. The infrared satellite image shows the complex cloud structure associated with a developing storm that created major flooding when it reached the coast. The flight track of the NOAA P-3 aircraft is shown (red for upper-levels, blue for lower levels). The low-level-jet (LLJ), which was observed to have winds exceeding hurricane force is highlighted. The measurement of a LLJ stronger than 70 knots was reported to the National Weather Service Forecast Office in Monterey, which then issued a flash flood warning that gave 6-h lead time for a record- breaking flash flood on Pescadero Creek (inset).

The strategy is based on applying research techniques and technology development over several years to address specific forecast problems, such as flood warnings, high wind warnings, and marine warnings. An ongoing series of focused field experiments is envisioned over the coming years to test new instruments, data assimilation techniques and operational tools that emerge from analysis of the previous experiments, and to develop a strategy to anticipate the nature of extreme weather events linked to climate variability (e.g., ENSO).

d) Societal impacts and forecast user input

• Forecasters, and forecast users on the U.S. West Coast placed improved prediction of coastal storms, including regional and local impacts, as a high priority problem in their region (Appendix 1)^#.
• Although operational stream flow models depend on QPF being provided in 6-h increments, and small streams respond even more quickly, 6-h QPF is much less accurate than 24-h QPF (Fig. 4).
• Hazardous weather in the West-Coast states of CA, OR, and WA due to these storms has caused major loss of life and property* (mostly due to flooding), and losses have increased in recent years:
• 37-year average of 9 fatalities per year; 26 fatalities per year during the last 4 years, and
• 15-year average of $700 million damage per year; $1,900 million/year over the last 4 y.
• These weather impacts are comparable to the annually averaged effects of earthquakes in this earthquake-prone region over the last 30 years, which included 4 major seismic events^@.
• The U. S. West Coast is exceptionally vulnerable to damage from such storms because:
• the region is characterized by high population densities and rapid population growth, and
• west coast weather prediction is impeded by a relative sparsity of data over the Pacific.
• CALJET and other research has shown that major forecast improvements are possible.
• Flooding is the most common and costly issue in emergency response in the area.
• Watershed impacts of landfalling storms dominate local and regional water supply.
• Because winter storms tend to move from west to east, positive effects of new data over the Eastern Pacific can spread eastward over time covering much of the U. S. over up to 4 days.

^#Input at PACJET Workshop 9/99.
*Statisitics: National Climatic Data Center
and ^@National Geophysical Data Ctr.

Fig. 4. Precipitation forecast threat scores from an NWS study of the QPF process. The assessment was made for the area covered by the California/Nevada River Forecast Center (CNRFC) using forecasts and observations of 6-h and 24-h accumulated rainfall based on numerical model output alone (the AVN and ETA models) and on the operational QPF issued by the forecast offices.

2. Linkages to national and NOAA priorities

• Creates a strategy for responding to seasonal-to-interannual forecasts of increased likelihood of severe coastal storms associated with the El Niño-Southern Oscillation (ENSO).

• Explores mesoscale physical processes that can locally amplify large-scale effects of climate variability.

• Focuses on identifying an optimal observing system for short-term (0-24 h) mesoscale QPF, which addresses two of USWRP's three core areas: quantitative precipitation forecasting, and studies of optimal observing systems for weather prediction.

• Accelerates development and field tests of new instruments and observing strategies for potential use during a large Pacific experiment (THORPEX) being considered by USWRP.

• Links USWRP objectives to the problem of coastal weather impacts and end-user needs, the importance of which is highlighted by NOAA's COASTS Initiative NOAA (Fig. 5).

• Addresses high priorities within the NWS strategic plan, including improved
• prediction of runoff through better QPF and NEXRAD quantitative precipitation estimation

• wind forecasts in the coastal zone, and

• prediction of orographic precipitation enhancement.

Fig. 5. Schematic summary of the COASTS initiative under development within NOAA.

3. The strategy for PACJET and for addressing West Coast short-term prediction

This section briefly summarizes the experimental goals, which attempt to balance basic and applied research with operational applications development.

a) Research:

• develop and test new instruments and observing strategies,
• test a buoy-mounted wind profiler
• implement a specialized scanning strategy for GOES to optimize feature-tracked winds

• evaluate vertical temperature, moisture, and wind fields derived from AMSU satellite data

• test an airborne scanning microwave radiometer (a candidate for future satellite application)

• field test UAVs in the difficult operational environment found over the Pacific in winter

• test the ensemble transform method for dropsonde targeting on the mesoscale

• use multiple aircraft to test hybrid sampling approach developed at the end of CALJET

• test the utility of guidance from medium-range to seasonal forecasts in staging P-3 missions

• explore how to best assimilate data to improve mesoscale forecast accuracy,
• quasi-real-time evaluation of data impact using parallel runs of operational models and data assimilation systems with and without PACJET experimental data

• examine predictability on the mesoscale by considering both the role of large-scale background features documented using Winter Storm Reconnaissance data from the central Pacific and smaller-scale features documented by PACJET in the eastern Pacific

• test the effect of different data assimilation techniques

• develop an ensemble of mesoscale models (RUC, ETA, MM5, COAMPS, ARPS)

• create experimental probabilistic QPF on the mesoscale and assess its performance

• explore optimal data assimilation techniques for improving mesoscale QPF

• develop tools to assimilate microwave satellite observations and airborne radar observations

• study physical processes and test parameterizations,
• extend collection of high-wind measurements of air-sea interaction

• compare ship and airborne methods used for flux measurements in high winds

• explore air-sea fluxes in drier inflow region of the warm sector

• measure extratropical warm orographic rain process found during CALJET

• measure 2-D wave spectrum using an airborne wave mapping system (Appendix 2).

b) Operational applications development:

• real-time assimilation of experimental data for operational NWP,
• assimilation into an operational global scale model (MRF)

• assimilation into an operational mesoscale model (experimental version of the RUC)

• creation of forecast tools around new instruments and observing strategies,
• develop data dissemination pathway for visual and quantitative experimental data

• create visualization products designed for optimal use of data for specific forecast problems

• initiate an alert to forecast offices that experimental data will be available

• develop real-time verification tools around special PACJET observations

• monitoring impact of experimental data on NWP and on issuance of warnings
• training forecasters and key forecast users on new data and concepts before PACJET

• tracking the direct use of new data by forecasters and users during the experiment.

4. PACJET and the linkage between climate variability and extreme weather events

Although it is becoming well established that large-scale climate variability such as ENSO can modulate seasonal-to-interannual variations in regional climate, in the case of extreme precipitation it is the specifics of each storm that determine many of the societal impacts of these climate anomalies. PACJET explores this linkage in the following ways:

• develops an observing strategy for improved short-term prediction of extreme west coast weather events that builds on advances in seasonal-to-interannual forecasts,
• the determination of the precise location of a highly instrumented coastal site for studies of coastal processes will be made based on guidance from 6-month lead time forecasts,
• two operations centers (a primary site and a secondary site) are built into the plan to allow the flight operations to be moved north or south in response to changes in storm track predicted in the 5-10 d forecasts,
• input from tropical and seasonal-to-interannual specialists will track the development and progression of Madden-Julian-Oscillation or other tropical forcing mechanisms that cause the eastward extension of the Pacific storm track, and thus leads to episodes of extreme coastal storms,
• explores the role of coastal sea surface temperature anomalies and ocean mixed layer heat content, which are linked to Pacific-basin oscillations such as the El Niño-Southern Oscillation (ENSO) and the Pacific Decadal Oscillation (PDO) in amplifying or suppressing coastal orographic rain, and flooding,
• observes the spatial variability of water vapor content associated with changes in storm track,
• and documents the mesoscale structure of coastal storms in a way that allows comparison with idealized studies of the dependence of storm structure on ENSO.

5. Data assimilation studies to improve short-term west coast weather prediction

The evaluation of data impact and the use of different models will allow investigation of several issues that are important to short-term QPF and wind prediction over the US west coast. First, it will enable comparisons to be made between forecasts/simulations from different models through verification against specialized field experiment data, so that the performance of each model in coastal wind and precipitation prediction will be evaluated. Second, the difference in the forecasts/simulations revealed by the evaluation of data impact will provide information on the sensitivity of coastal predictions to perturbations in the model physics and initial/boundary conditions. Third, it will provide a unique opportunity to test the concept of mesoscale ensemble forecast for coastal weather prediction. Last, it provides a test bed for future improvement and development of an optimal data assimilation strategy for coastal weather prediction.

Data assimilation for coastal prediction is extremely challenging because conventional observations are sparse in the eastern Pacific. The initialization of a model for very short-range prediction over the US west coast requires optimal assimilation of information from unconventional data sets, such as those from satellite and targeted observation missions. There are several techniques available for assimilating unconventional data, but each technique has its advantages and disadvantages that vary with different operational settings and instrumentation. Thus, the variety of instruments that will be applied in PACJET makes it a valuable opportunity to tackle issues regarding the superiority of one technique over the other for short-range prediction over the US west coast, and to reveal the statistics and sensitivity that are critical for optimizing data assimilation.

A few specific approaches that are envisioned include:

Quasi-real-time evaluation of data impact using operational models and NWS verification

• synoptic scale evaluation using a global-scale model
• mesoscale evaluation using an experimental version of the RUC optimized for PACJET

Real-time use of experimental models

• develop an ensemble of mesoscale models (RUC, ETA, MM5, COAMPS, ARPS), including variations in their associated data assimilation systems
• develop and test improved data assimilation techniques (3-D multiquadric, 3DVAR, 4DVAR, etc...) and time-varying estimation of forecast error magnitude and spatial correlation (some can be done in real-time parallel runs, but most will occur later).

Post-experiment research on optimal data assimilation techniques for mesoscale QPF and probabilistic QPF

• evaluate QPF impact of various observing systems using MM5, COAMPS, and RUC models with various assimilation schemes such as multiquadric interpolation, 3DVAR etc
• assimilation of passive microwave satellite observations

Predictability on the mesoscale

• consider both the role of large-scale background features documented using Winter Storm Reconnaissance data from the central Pacific and smaller-scale features documented by PACJET in the eastern Pacific
• explore the scale of sensitivity regions through data impact studies exploring the effects of sub-sampling high-density PACJET data.

6. Status of the PACJET experiment as of January 2000

• Planning Workshop held in September, 1999 gathered input from researchers, NWS forecasters, emergency managers, water managers, the fishing industry, and others.
• Set of recommendations prepared based on Planning Workshop.
• Request submitted for two NOAA P-3 aircraft and the Ron Brown, a new research vessel.
• Extensive upper-air and land-processes network committed to CA for air quality research (Fig. 6).
• Coastal wind profilers, cloud radars, and mobile balloon soundings from NOAA/ETL.
• Key Briefings: NWS/NESDIS/OAR strategic planning group (Aug. '99), USWRP Scientific Steering Committee (Oct. '99), NOAA COASTS Initiative Planning team (Oct. '99).
• A PACJET Technical and Scientific Planning Workshop will be organized for early 2000.

Click to view full sized figure.

Fig. 6. Base map showing combined PACJET/air quality study.

7. Participants

This list of organizations is based mostly on participation at the PACJET Planning Workshop that was conducted in Sept. 1999 in Monterey, CA, and represents a core group likely to participate in PACJET.

Science and operations groups:

• NOAA Research Laboratories (Environmental Technology Lab., National Severe Storms Lab., Forecast Systems Lab.)
• NOAA/NWS/Western Region Weather Forecast Offices and River Forecast Centers.
• NOAA/NCEP Centers for Environmental Modeling, Hydrometeorological. Prediction, and Marine Prediction
• NOAA/NESDIS CIRA and CIMSS
• Naval Postgraduate School
• Naval Research Laboratory
• University of Nevada at Reno, Desert Research Institute
• University of Colorado/Cooperative Institute for Research in Environmental Sciences
• SCRIPPS/Joint Institute for Marine Observations
• University of Oklahoma
• Northwest Research Associates

Forecast user groups:

• California Governor's Office of Emergency Services
• Pacific Coast Federation of Fisherman's Associations
• California Department of Water Resources
• Broadcast meteorologists
• Consulting meteorologists

Mesoscale modeling groups who have indicated likely involvement in PACJET:

• ETA: NCEP/EMC
• RUC: Forecast Systems Laboratory and NCEP
• MM5: Naval Postgraduate School, California Mesoscale Modeling Consortium, NOAA/ETL
• COAMPS: Naval Research Laboratory and Naval Postgraduate School

8. Payoffs of PACJET and a long-term effort to improve short-range west coast forecasting

• Creation of an observing strategy for improved short-term prediction of extreme west coast weather events that builds on advances in seasonal-to-interannual forecasts
• Enhanced protection of life and property due to better weather warnings on U.S. west coast.
• Better forecasts over much of the U. S. as coastal storms then move east across the continent.
• Enhanced West Coast emergency response through use of new data in decision making.
• More effective reservoir operation, with improved flood control and water conservation.
• Better dissemination of weather warnings via the web, television and other media.
• Given the large annually averaged losses due to storms in the West Coast states (9-26 deaths and $0.7-1.9 billion in damage), even small forecast improvements could have great impact.

Appendix 1

Workshop to Explore Applications of a Field Experiment (PacJet)

to 0-24 h West Coast Winter Weather Forecasting

Background

Experience gained during the recent California Land-falling Jets Experiment (CALJET) indicates that real-time use of experimental data from the coast to 1000 km offshore can aid in issuing coastal weather warnings. In particular, the Monterey forecast office of the National Weather Service (NWS) used CALJET research aircraft measurements of the strength of a low-level jet in an approaching storm to help issue a flash flood warning that gave 6 h lead time for a record-breaking flash flood in the region. It has also been shown recently by NCEP that insertion of CALJET aircraft data in storms roughly 24 h before they made land-fall led to improvement in the 24-h operational numerical weather predictions in all of the 5 cases tested. In addition, operational and research profiler networks on the U. S. west coast are now providing real-time monitoring of the coastal environment, albeit fewer are now available than in recent years. In particular, coastal wind profilers provide a real-time view of weather phenomena, such as blocking and barrier jets along the coast that can modify the location of precipitation maxima.

These results, along with other research, have motivated the preliminary design of a follow-on experiment referred to as PacJet. A planning letter has been submitted for use of two NOAA research aircraft and NOAA's state-of-the-art research vessel (the Ron Brown) off the west coast of the United States during the winter of 2000/2001. In addition, data from numerous wind profilers that will be deployed in this region for air quality research will also be available for use. Although the experimental design will be modeled after our 1998 El Niño experience, the location of the experiment may be adjusted depending on threats assessments provided several months ahead of time based on seasonal-to-interannual forecasts (i.e., El Niño and La Niña). The value of seasonal-to-interannual forecasts and forecast uncertainty will also be discussed.

We invite your participation in a workshop aimed to provide recommendations from the operational weather forecasting community along the US west coast that can aid directly in the detailed design of the field program. The emphasis will be on 0-24 h forecasting of land-falling Pacific winter storms, covering the area from roughly 200 km inland to 1000 km offshore. Illustrations of atmospheric, hydrologic, and ocean sea state forecasting issues are encouraged. This input, combined with the basic research objectives, will set the priorities for the experiment. The concept for this meeting emerged from a joint session of the CALJET Workshop and the National Weather Service Western Region Marine Forecasters Training Course in May 1999 where marine forecasters indicated that such data could be very useful, and provided several novel recommendations on what would be useful and why.

The experimental objectives identified thus far have been developed out of priorities established in several reports by NOAA, the National Weather Service, the U.S. Weather Research Program (USWRP), and the National Research Council which include research aimed at:

• improved quantitative precipitation forecasting,
• studies addressing the optimal mix of observations for weather prediction,
• effects of topography on local weather regimes,
• design of a Pacific Coastal Forecast System, and
• regional assessment of seasonal to inter-annual forecasts.

Meeting objectives

• Familiarize forecast community with recent field programs and the preliminary design for PacJet.
• Develop a set of recommendations from the forecasting community for use in designing PacJet, with emphasis on needs and opportunities for real-time use of experimental data, and methodologies for evaluating impact/utility of the experimental data in real-time applications. Assimilation of the experimental data into operational numerical models will also be discussed.
• Decide on approaches to enable operational forecasters to have real-time access to these data.

____________________________________________________________________________________

____________________________________________________________________________________

Tentative Agenda

Session A: Introduction

Summary of experience gained from CALJET and two recent central Pacific observing programs. Description of preliminary design for PacJet and how input from this meeting can affect the experiment.

Session B: Illustrate forecast problems and needs

To bring out the most important areas where experimental data could aid in operational forecasting, presentation of case studies by the forecasting community are requested. These case studies are intended to illustrate where additional coastal, inland, and offshore data (up to 1000 km offshore) could be of most value. These examples should focus on the winter season along the U. S. west coast, on the region from roughly 200 km inland to 1000 km offshore, and on the 0-24 h forecast time window. Illustrations of atmospheric, hydrologic, and ocean sea state forecasting issues are encouraged. It is anticipated that each forecaster or Science Operations Officer would use about 20-30 min to illustrate their region's forecast problems. This is enough time to see in depth what the problems are and to discuss them. This is the heart of the workshop and will form the basis for the recommendations developed the next morning in session C. See http://www7.etl.noaa.gov/programs/CALJET/index.html for a description of CALJET.

Session C: Develop set of recommendations

Combine input from the forecasting community with that from the research team concerning the capabilities and limitations of new observing systems and strategies, to develop a detailed set of recommendations for use in designing PacJet.

____________________________________________________________________________________

____________________________________________________________________________________

Final Agenda

Tuesday 31 August 1999

Session A: CALJET Experience and PACJET Planning

Session Chair: Warren Blier, Science and Operations Officer, NWSFO Monterey, CA

9:00 AM Introduction and meeting objectives

9:05 AM "What was CALJET and what have we learned from it?" (Ralph)

9:25 AM "Can targeted dropsonde deployments improve coastal storm forecasts?" (Toth)

9:45 AM "PACJET: Preliminary plans, candidate observing systems, and status" (Ralph)

10:00 AM "Special PacJet Satellite Products" (Velden/Ralph)

10:10 AM "Studying Surface Wind Fields in Oceanic Storms with SSM/I, Scatterometer, and SAR data (Walter)

10:20 AM "Geostationary and Polar Orbiting Satellite Data during PacJet" (Dostalek)

10:35-11:00 AM Break

11:00 AM "RUC Mesoscale Modeling Runs for PacJet" (Benjamin)

11:15 AM "Quasi-operational mesoscale modeling in California: Application to PacJet" (Miller/Schmidt)

11:30 AM "Sea state prediction with coupled mesoscale models: PacJet as a test bed?" (Wilczak)

11:45 AM "NOAA's Coastal Forecast System and SAFESEAS" (Contorno)

12:00 Noon "The THORPEX Experiment" (Langland)

Session B: Illustrate forecast problems and needs

Session Chair: Brad Colman, Science and Operations Officer, NWSFO, Seattle, WA

1:20 PM "The NWS Quantitative Precipitation Forecast Process Assessment" (Reynolds)

1:40 PM "NCEP/Hydrometeorological Prediction Center QPF Skill and its Relationship to Numerical Weather Prediction, Can PacJet Improve the Models?" (Reynolds)

2:00 PM "How CNRFC fits into the QPF process and how we use the QPF" (Rhea/Strem)

2:20 PM "How the Sacramento Flood Control System Works," (Roos)

2:40 PM "NCEP Marine Prediction Center Overview of Operations and Critical Forecast Problems" (Partain)

3:00-3:45 PM Break

3:45 PM "Impacts of Winter Season Pacific Storms on Southern California" Oxnard (Danielson)

4:00 PM "Perspectives from the Hanford NWS Forecast office" (Greiss)

4:15 PM "Winter Hydrometeorological Concerns at WFO Sacramento" (Cunningham)

4:30 PM "Perspectives from the Monterey NWS Forecast Office" (Blier)

4:45 - 5:15 PM Discussion "Needs and opportunities for improvement"

5:15 PM Session B ends for the day. Session B continues Wednesday morning.

______________________________________________________________________________

Wednesday 1 September

Session B Continued

Session Chair: Wendell Nuss and Doug Miller, Naval Postgraduate School, Monterey, CA

8:30 AM "Perspectives from the Eureka NWS Forecast Office" (Nordquist)

8:45 AM "Perspectives from the Seattle NWS Forecast Office" (Colman)

9:00 AM Tom Maruyama, San Mateo County Emergency Management

9:15 AM Neal O'Haire, California Governor's Office of Emergency Services, Coastal Region

9:30 AM "Perspectives from a Television Broadcast Meteorologist" (Lynn)

9:45-10:15 AM Break

Session C: Synthesis and Recommendations

Session Chair: Marty Ralph, NOAA/Environmental Technology Laboratory, Boulder, CO

10:15 AM - 12:00 Noon Discussion

What are the key forecast issues?

What experimental data would be most useful?

How should the data be made available?

Should training be considered just before the experiment?

How to best assess operational impact, both NWP and real-time?

Noon Workshop ends.

Appendix 2

A Scanning Radar Altimeter for use on a P-3 During PACJET

The Scanning Radar Altimeter (SRA) [1, 2] and its predecessor, the Surface Contour Radar (SCR) [3, 4] were designed primarily to measure the energetic portion of the directional wave spectrum by generating a topographic map of the sea surface. The measurement geometry is shown in Figure A.1A. The SRA sweeps a radar beam of 1^o half_power width (two_way) across the aircraft ground track from _22^o (off_nadir to the left side of the aircraft) to +22^o (off_nadir to the right side of the aircraft), simultaneously measuring the backscattered power at its 36 GHz (8.3 mm) operating frequency and the range to the sea surface at 64 points spaced across the swath at 0.7^o incidence angle intervals. These slant ranges are multiplied by the cosine of the incidence angles (including the effect of aircraft roll attitude) to determine the vertical distance from the aircraft to the sea surface. Subtracting these distances from the aircraft height produces the sea surface height. The SRA produced these raster scan lines of wave topography at the rate of 10 Hz.

Figure A.1B shows a surface elevation profile extracted from the middle of the SRA swath during a flight into hurricane Bonnie on 24 August 98 at 1.5 km altitude. The crest-to-trough height of the highest wave is almost 18 m. This wave was the highest in the 10 km segment of grey-scale coded SRA wave topography data shown in Figure A.1C. The SRA swath width was 1.2 km and the wave image is in proportion. The aircraft was traveling toward the north and the image shows the general spatial variation the waves in the predominantly northwest propagating swell wave field 150 km north of the hurricane Bonnie eye.

Figure A.1D shows the directional wave spectrum generated from the SRA topography of Figure A.1C. The significant wave height (SWH) was 9.1 m. The nine contours are linearly distributed between 10% and 90% of the peak spectral density. The three solid circles indicate wavelengths of 100, 200, and 300 m. This spectrum was quite narrow and its peak corresponded to a 300 m wavelength wave propagating toward 330^o. The long arrow indicates the wind at that location, which was blowing toward 258^o at 40 m/s. There were wind-driven waves propagating in the downwind direction, but their energy density was below the 10% level of the spectral peak. Figure A.1F shows a 10-km segment of gray_scale-coded wave topography measured while the aircraft was traveling toward the southeast about 30 km south and 40 km east of the eye. The image has been rotated about 45^o clockwise from its original track orientation. The time sequence of the data progresses from top to bottom.

Figure A.1E shows the directional wave spectrum generated from the SRA data of Figure A.1F. The SWH was 6.7 m and the wave system was bimodal with approximately the same energy in waves propagating almost at right angles to each other. The wind at this location was also about 40 m/s, but toward 22^o. Waves of about 140 m wavelength propagated toward 28^o, nearly aligned with the wind. The spectral peak of the swell system corresponded to waves of 200 m wavelength propagating toward 315^o.

REFERENCES

[1] Walsh, E. J., L. K. Shay, H.C. Graber, A.Guillaume, D. Vandemark, D. E. Hines, R. N. Swift, and J. F. Scott, "Observations of Surface Wave_Current Interaction During SWADE," The Global Atmosphere and Ocean System, 5, 99_124, Gordon and Breach Science Publishers SA, 1996.

[2] Parsons, C., and Walsh, E.J., "Off_Nadir Altimetry," IEEE Trans. Geosci. Remote Sens., 27, 215_224, 1989.

[3] Walsh, E. J., D. W. Hancock, D. E. Hines, R. N. Swift, and J. F. Scott, "An Observation of the Directional Wave Spectrum Evolution from Shoreline to Fully Developed," J. Phys. Oceanogr., 19(5), 670_690, 1989.

[4] Walsh, E. J., D. W. Hancock, D. E. Hines, R. N. Swift, and J. F. Scott, "Directional Wave Spectra Measured with the Surface Contour Radar," J. Phys. Oceanogr., 15(5), 566_592, 1985.